Search

The concept of a semiconductor laser with a circular resonator, its advantages and particular problems are discussed. The pillbox resonator is introduced and its operation on whispering gallery modes is illustrated using a computer model. The experimental evidence of the guiding mechanism is shown, leading to the first demonstration of continous wave operation in a semiconductor ring laser with a threshold current of 24mA. The parameters of the GaAs/AlGaAs material that are relevant for the low threshold current operation are presented and all aspects of the fabrication procedure are covered, emphasizing the processes that led to smooth sidewalls and the low loss circular cavity. A further reduction of the threshold current to 12.5mA is shown, which is owing to a coating of silicon nitride that supresses the non-radiative recombination current and reduces the scattering loss. The excess bending loss is calculated to be 3dB/360 and found to be independent of the radius between 30mum and 145mum. The influence of the Y-junction on the operation characteristic is studied and shown to cause kinks in the L-I curve; it is also held responsible for the relatively low differential quantum efficiency (0.02-0.04) of the devices. Strip-loaded guiding is demonstrated for radii between 300mum and 600mum and proposed as a solution for the problem of degradation that is caused by etching through the active layer. The integration capability of the structure is demonstrated by the succesful operation of a circuit comprising of a ring laser, a low-loss waveguide and a detector, and an optoelectronic integrated circuit featuring a ring laser and a field-effect transistor. The material parameters that are involved in performing these complicated functions are discussed and modeled numerically.

A major goal of nanophotonics research is to enhance light-matter interaction and to develop novel functionalities. The work in this thesis addresses this goal in 2 areas, i.e. microscopy and laser emission, using guided mode resonances (GMRs) as a tool. GMRs exhibit significant field enhancements, which makes them interesting for the enhancement of light-matter interactions, such as light scattering or fluorescence emission; the challenge was to show whether and how GMRs could be used to improve microscopy methods such as interferometric scattering microscopy (iSCAT) and fluorescence microscopy. The second challenge was to show whether GMRs could be used to enhance laser operation of 2D-material lasers. In the first part of the thesis, I describe how I built, tested and optimised an optical system that allows the study of enhanced microscopy. Then I demonstrate an efficient method for quantifying the enhancement of fluorescence excitation with a chirped GMR grating. With regular iSCAT, I was able to detect gold nanoparticles as small as 20 nm. In the second part, I show how a two-dimensional GMR grating can be used to both enhance the pump absorption and the laser emission of a WS2 monolayer. The devices, lasing with a low pump threshold, have a much higher output power and spatial coherence than other 2D-material lasers reported in the literature. The contrast enhancement in microscopy opens new possibilities, as it can either push the detection limits or simplify the setup requirements for detecting a given particle. Regarding the lasing application, I showed that the concept of GMR-based lasers with large area 2D-material gain media is promising. The concept could be adapted to other 2D-materials emitting at wavelengths relevant for communication or sensing applications. This would represent an important step towards operation with reduced power consumption or for opening novel opportunities for making lasers.

Photonic sensors have the potential to enable highly sensitive point-of-care diagnostic technology and become an essential element of the global health system. To maximise its benefits, the technology needs to provide affordability in addition to accuracy, which is a challenge that is reflected in the currently available diagnostic technologies. For example, the gold standard diagnostic technology, enzyme-linked immunosorbent assay, requires trained personnel and a well-equipped laboratory. In contrast, the low-cost lateral flow devices offer limited sensitivity. New technologies are emerging, such as label-free photonic biosensors that are intrinsically simple, as they can directly monitor the presence of a target biomarker without requiring additional processing steps. However, when high sensitivity is required, they typically rely on expensive components. Circumventing the connection between high cost and high performance in photonic biosensing and demonstrating a viable alternative requires a novel approach and is the primary aim of my thesis. To this end, I have developed a common-path interferometric sensor based on guided-mode resonances to combine high performance with inherent stability. The concept of spatially superimposing two beams carrying the phase information of two orthogonally polarised resonant modes and interferometrically measuring the relative phase difference between them is here introduced and experimentally realised. To validate the developed technology, the sensor is applied to bulk refractive index sensing as well as the proof-of-principle detection of procalcitonin. The results of this thesis reveal the interesting physics involved in an interferometric probing of resonances in dielectric nanostructures and indicate the potential of the developed sensor for high-performance diagnostics based on photonics.

Personalised healthcare, local environmental monitoring and food quality control drive the need for compact, cost-effective, reliable methods for detecting specific molecules. Resonant refractive index sensors can detect these molecules binding to their surfaces with label-free measurements. The resonators have a small device footprint which is ideal for miniaturisation and they are easily integrated with electronic systems through ubiquitous optoelectronic components. These resonant refractive index sensors have evanescent fields which probe the detection layer. Their sensitivity is often optimised by maximising bulk sensitivity, which favours spatially extended modes. This strategy is not necessarily suitable for maximising biomolecule detection. The more relevant consideration is the sensitivity to changes within a molecule-length of the surface. In this work, I show that high surface sensitivity can be achieved by tightly confined resonant modes in metal-insulator-metal (MIM) structures, which have relatively high surface sensitivity despite low bulk sensitivity compared to other resonant modes. I achieve 55 nm/RIU sensitivity to a 10 nm layer with lithographically fabricated arrays of third-order MIM resonators, approximately twice as high as that of an extended SPP mode also supported by the periodic array. I increase the MIM surface sensitivity by accessing the first order MIM mode with bottom-up methods and estimate 65nm/RIU sensitivity from displacing a ~3nm polymer capping layer. I complement the nanocube MIM resonators with a readout assembly using low-cost optoelectronic components which achieves a 5.7x10⁻⁴ RIU or 3.7x10⁻³ RIU bulk limit of detection when controlled by either research-grade or lowcost electronics respectively. The simulations and experiments in this work emphasise the importance of surface sensitivity as a performance metric. The nanocube structures present a powerful sensing modality which achieves high surface sensitivity and forms an integral part of the low cost hand-held sensing assembly. These design considerations and prototypes inform the next generation of optical biosensors.

Early diagnosis and targeted disease treatment are essential elements of healthcare provision. Current systems often revolve around symptomatic diagnosis and laboratory testing when needed. These systems often result in semi-blind treatment or long wait times. Point-of-care biosensors that can quantify specific biomarkers have an important role to play in this context as they bridge the gap of quick and specific testing in healthcare provision. Photonic biosensors have been demonstrated as a laboratory diagnostic tool with well-established advantages, achieving low limits of detections for protein biomarkers. Several issues remain before they can be translated into commonly used point-of-care instruments. Ideally, integrated photonic biosensors should be inexpensive, sensitive and easy to use. However, many existing photonic biosensors require complex readout equipment, such as external spectrometers, or precise and bulky optical coupling. Presented in this thesis is the development of a low-cost photonic biosensing instrument. The modality used is that of a chirped one-dimensional guided mode resonance sensor. When used in conjunction with a monochromatic source, detection can be accomplished simply by a camera in a compact microscope-style configuration. A study into the performance of low-cost components was conducted to understand the sources of noise. With the understanding gained from this study the instrument was able to achieve a bulk refractive index sensitivity of 3.1 ±0.6 ×10-5 Refractive Index Units. This limit of detection is comparable to other laboratory based modalities and so the device was tested with C-Reactive Protein and Immunoglobulin G assays as example proteins. The best performance achieved was detection of 1 ng/mL for C Reactive Protein, well below the clinically relevant range. The limits of detection achieved using low-cost components indicates that photonic biosensors are well suited to clinical situations and have the potential to play a significant role in the development of diagnostics in the future for healthcare.

Biosensing technology currently uses a single transduction mode, restricting the systems the technology can be applied to and the information that can be gained. Multimodal biosensing combines multiple transduction technologies to probe different properties simultaneously, increasing the range of measurable interactions, the amount of information that can be extracted, and the detection accuracy. Guided mode resonance sensors are one dimensional periodic grating structures that measure the local refractive index at the grating surface, and have been used extensively for label-free biosensing assays. Electrochemical biosensors measure the activity of chemical reactions through changes in electrical current. Combining refractive index and electrochemical sensing allows for the parallel interrogation of the presence and activity of biomolecules. A multimodal electrochemical guided mode resonance (EGMR) sensor based on a 1D GMR grating structure with a layer of indium tin oxide is presented. Simulation of the GMR structure was used to inform the EGMR design for the best compromise between electrochemical and optical sensing performance. The fabricated device is characterised optically and electrically. Combined electrochemical and optical sensing measurements were demonstrated in parallel to characterise a redox active molecule. The electrochemical chirped GMR (ECGMR) was applied to developing a low cost multiplexed label-free biosensor. The chirped GMR sensor converts the spectral response of the GMR to spatial information allowing binding to be measured using a monochromatic source and a camera. A biocompatible antibody electrografting protocol was demonstrated for creating a multiplexed antibody array of sepsis biomarkers using the ECGMR. Parallel label-free detection of CRP, IgG and Escherichia coli was demonstrated. This work is the first example of a multimodal GMR, and progresses the use of biosensing technology for point-of-care applications.

Microprocessor clock speeds have stagnated in the 2010s no longer keeping up with Moore's Law. With node sizes down to 7 nm it is becoming difficult to fit more transistors onto a chip due to quantum tunneling. The energy consumption of microprocessors is also a major problem for climate change and the battery life of mobile devices. To combat these problems we look to silicon photonics to provide a solution. Transmitting data optically has two main benefits, reducing the resistive heating of electrical data transfer and increasing possible data transfer bandwidth. These two properties in concert would allow a microprocessor with optical data connects to run faster and use less energy. The main problem with silicon photonics is the lack of an on-chip light source due to the poor light emission of silicon. An on-chip light source must be compatible with current CMOS fabrication processes and easy to mass produce. We seek to solve this problem by employing two-dimensional transition metal dichalcogenides (TMDs) as a laser gain material. Two-dimensional materials became a popular area of research after the "graphene revolution" pioneered by Andre Geim and Konstantin Novoselov. The techniques developed to produce atomically thin layers of graphene can be transferred to TMDs. TMDs exhibit the same strong in-plane bonding and weak interlayer van der Waals bonding as graphene, however when thinned down to monolayers they have a direct bandgap. The direct bandgap makes them good optical emitters. In this thesis I review the performance of TMD microcavity emitters in the literature and then focus on two main areas, the design and fabrication of photonic crystal cavities to produce a laser resonator and the optical and structural characterisation of 2H-MoTe2. For technological applications, large area monolayers of 2H-MoTe2 that can be grown on a device are required. I investigate the optical and structural differences between 2H-MoTe2 exfoliated from a bulk crystal and the CVD growth of MoTe2.

Antimicrobial Resistance (AMR) is an important societal and medical burden being faced by every healthcare system in the world. The large overuse of antibiotics favours bacterial resistance, thereby causing drugs to become ineffective. While traditional susceptibility tests are well established, they are slow at informing prescriptions, which means that many antibiotics are prescribed without proper diagnostics. This inefficacy stems from the detection of bacterial growth at the bulk colony level. In addition, averaging over bulk colonies masks any cell-to-cell difference even though it is well known that bacterial populations and their response to antibiotics are heterogenous. These differences need to be considered to correctly assess the efficacy of any microbial agent in a timely manner. This thesis presents a multiparameter approach for profiling the susceptibility of individual bacteria to antibiotics. Hydrodynamic trapping provides the mechanism for retaining and examining hundreds of single E. coli in a microfluidic channel. The multiparameter aspect consists of the simultaneous assessment of bacterial motility and morphology upon exposure to antibiotics. This combined approach allows us to detect susceptibility and resistance of E. coli MG1655 in as little as 1 hour from spiking the culture. Standard microdilution assays and bacteria counts confirmed the validity of the classification. Additionally, the richness of single-cell data enables us to study the dynamics of population killing and the mode of action of the antibiotics. The methodological approach presented here has the potential to complement traditional susceptibility tests by providing a deeper understanding of the drugs' action and a more rapid way of assessing the bacterial response to antimicrobial agents. More generally, the assay provides a platform for monitoring the behaviour of hundreds of single bacteria over time, while preserving the individuality of each microorganism.

This thesis addresses two of the main materials for solar cells, namely silicon and the family of halide perovskites. For silicon, light trapping structures are investigated for solar cell applications while perovskite materials are investigated as a gain material for optoelectronic applications. Light trapping allows the capture of photons that might otherwise be lost, especially at the red end of the spectrum where silicon is less absorptive. The key is to enhance the efficiency of silicon cells by thinning down the wafer and reducing the bulk recombination losses and to achieve a higher Voc while maintaining strong light absorption (represented by a high short circuit current, Jsc) by applying efficient light trapping schemes. It is still an open question whether nanostructures are beneficial for real devices, especially since highly efficient solar cells employ >100 μm thick absorber materials and use wet etched micron-sized pyramids for light trapping. In this thesis, I conduct a study which compares nanostructures and pyramid microstructures on wafer-based silicon solar cells. This study is important because (1) most light trapping nanostructures are investigated only in the optical regime, while I realize them on silicon devices to analyze both their optical and their electrical character; (2) nanostructures perform better than microstructures in wafer based silicon solar cells, highlighting the effectivity of nanostructures even in wafer-based silicon. Here, the nanostructures comprise wet and dry-etched quasi-random structres and they are compared with pyramidal microstructures. A photocurrent as high as 38 mA/cm2 for a dry etched quasirandom nanostructure is attained experimentally, which is 3.2 mA/cm2 higher than wet etched pyramids fabricated in the same batch. The other material which is now becoming very popular in the solar cell community is the family of metal halide perovskite materials that are increasingly attracting the attention of optoelectronics researchers, both for solar cell and for light emission applications. The ultimate is in simplicity, however, is to observe lasing from a continuous thin film, which has not been aimed before. Here, I show perovskite random lasers; they are deposited at room temperature on unpatterned glass substrates and they exhibit a minimum threshold value of 10 μJ/cm2. A rather special feature is that some of the films exhibit single and dual mode lasing action, which is rarely observed in random lasers. This work fully exploits the simplicity of the solution-based process and thereby adds an important capability to the emerging field of perovskite-based light emitters.

With the advent of techniques such as Stimulated Emission Depletion Microscopy (STED) and light sheet microscopy, the generation of specialised light beams has become an exciting field. However, while very effective methods of generating such beams exist, the components necessary to do so are generally large and cumbersome. Metasurfaces promise the replacement of these traditional bulky optical elements with sub-wavelength thick and flat alternatives, paving the way for integration into microscale form factors. Metasurfaces commonly use a distribution of nanoscale resonant elements to engineer a phase plate that shapes light through the Huygens' principle, allowing them to mimic and improve upon traditional optics. Initially, plasmonic resonant elements were explored by the community, but their dissipative losses have severely limited the efficiency of these devices. Here I discuss my work on the development of dielectric sub-wavelength grating based metasurfaces. Four types of metasurface, each using a different manifestation of grating physics are explored: direct phase, polarisation conversion, geometric phase, and active metasurfaces. I show that these different types of metasurface together allow the shaping of a wide variety of beams under a large range of different conditions, while retaining efficiencies on the order of 80-90%. Examples of the beam shapes explored include focused beams, vortex beams, Bessel beams and cylindrical vector beams. The development of high efficiency dielectric metasurfaces brings ultrathin optics closer to practical applications. Their materials and sizes facilitate the integration into previously unavailable form factors, including applications in microfluidics.

Our increased understanding of the molecular biology of disease has had a significant impact on healthcare. Genetics has allowed the identification of hereditary diseases and predisposition for others, such as cancer. However, for many diseases is it also necessary to monitor the expression of panels of proteins. The need to monitor protein expression presents a significant technological challenge requiring a highly multiplexed analytical technology with a sensitivity down to femtomolar-range. Low-cost photonic devices are highly sensitive to changes in their local environment and can be chemically modified to exhibit high specificity detection towards, for instance, proteins or DNA oligonucleotides. However, even if small-footprint photonic biosensors can be engineered in silicon microarrays, approaches to realise the very high-density, multiplexed sensing potential of photonic biosensors have yet to be demonstrated. This study aims to develop and demonstrate, for the first time, a dual-mode electro-photonic technology capable of highly multiplexed detection at the submicron scale and multiparameter profiling of biomolecules on the silicon photonics platform. Furthermore, the technology integrates electrochemical and photonic measurements in a single sensor platform. By combining the complementary information revealed by each of the domains it is possible to broaden the range of systems that are accessible for silicon photonics. Our dual-mode technology consists of microring resonators optimally n-doped (doping density of 7.5 x 10 16 cm −3 ) to support high-Q resonances (Q-factor ≈ 50, 000) alongside electrochemical processes in situ. This combination of sensing mechanisms enables the application of electrochemical methods for site-controlled immobilisation of receptor molecules. Furthermore, electrochemical characterisation of molecules bound to the sensor surface also provides direct quantification of binding density and unique insight into chemical reactivity, which is unavailable with photonic detection alone. This unique technology, based on the combination of electrochemical and photonic sensing on a silicon platform, not only enables detection of multiple biological molecules required for future clinical diagnostics, but also has the potential to impact on fundamental biochemical research.

Optical biosensors make up a valuable toolkit for label-free biosensing. This thesis presents a detailed study on resonant grating surfaces for biosensing. The focus is on silicon nitride gratings, which exhibit a guided-mode resonance that is highly sensitive to refractive index variations in the vicinity of the grating. A sensitivity of 143 nm/RIU (refractive index units) is measured, leading to a detection limit of 2.4×10−4 RIU. This performance is shown to be sufficient for the detection of biomolecular binding down to ng/mL concentrations. With out-of-plane excitation, these gratings can be used as a sensing surface, enabling a spatially-resolved measurement of variations in refractive index; resonance imaging. The minimum detection distance (sensing depth) is measured to be 183 nm away from the grat- ing, while the spatial resolution of resonance imaging is found to be asymmetric: 2 μm parallel to, or 6 μm perpendicular to the gratings. Using a novel approach of fabricating a resolution test pattern on top of the grating, the relationship between resolution and index contrast is studied - an important question in the context of biosensing - where it is found to decrease with index contrast. All experimental results are supplemented with theoretical and computational models. The resonant gratings are then extensively applied to the study of biofilm development, cellular imaging, and the imaging of cellular secretion. Finally, a miniaturised biosensor is demonstrated, based on a chirped resonant grating. By tuning the resonance wavelength spatially on the chip, the resonance information is directly translated into spatial informa- tion. Instrument read-out requires just a monochromatic light source and a simple CCD camera, resulting in a final device that is inexpensive, compact, robust and can be remotely operated. Performance is proven with successful detection of biomolecular binding.

The manipulation of photons, especially the control of spontaneous emission, has become a core area of photonics research in the 21st century. One of the key challenges is the control of the broadband emission profile of thermal emitters. Recently, attention has focused on resonant nanophotonic structures to control the thermal emission with most of the work concentrating on the mid-infrared wavelength range and/or based on metallic nanostructures. However, the realisation of a high temperature, single wavelength, narrowband and efficient thermal source, remains a challenge. In this project, four individual nanophotonic resonant structures are presented for the control of thermal emission, all operating in the near-infrared (≈ 1.5 μm) wavelength range. The work is split over two different emission materials; gold and doped silicon. While I present two successful designs of narrowband thermal emitters from gold, the main backbone of the research is concentrated on doped silicon as the emission material. By combining the weak broadband absorption of doped silicon with a photonic crystal resonator, resonantly enhanced narrowband absorption is achieved. Using Kirchhoff's law of thermal radiation which equates the absorptivity and emissivity, narrowband absorption leads to narrowband emission, which is the underlying principle used throughout the work presented in this thesis to achieve narrowband thermal emission. One common oversight in many of the presented thermal emitter designs is the angular emission dependence, i.e. how the emission wavelength behaves away from surface normal. Typically, since the majority of the devices are based on periodic structures, the resonant emission wavelength changes with emission angle, which is not ideal. Here, the angular sensitivity is considered and addressed, by constructing a device that is based on localised confined resonances and not on propagating resonances, it is possible to achieve a truly monochromatic source i.e. one with the same emission wavelength in all directions, all the way up to an angle of 90°. Finally, the devices presented here demonstrate that weak absorption together with photonic resonances can be used as a wavelength-selection mechanism for thermal emitters, both for the enhancement and the suppression of emission away from the resonant wavelength.

Thin-film silicon solar cells have the potential to convert sunlight into electricity at high efficiency, low cost and without generating pollutants. However, they need to become more competitive with conventional energy technologies by increasing their efficiency. One of the key efficiency limitations of using thin silicon absorber materials relates to the optical loss of low-energy photons, because the absorption coefficient of silicon decreases strongly for these low-energy photons in the red and near-infrared, such that the absorption length becomes longer than the absorber layer thickness. If, in contrast, the incident light was redirected and trapped into the plane of the silicon slab, a thin-film could absorb as much light as a thick layer. Diffractive textures can not only efficiently scatter the low-energy photons, but are also able to suppress the reflection of the incident sunlight. In order to take advantage of the full benefits that textures can offer, I outline a simple layer transfer technique that allows the structuring of a thin-film independently from both sides, and use absorption measurements to show that structuring on both sides is favourable compared to structuring on one side only. I also introduce a figure-of-merit that can objectively and quantitatively assess the benefit of the structuring itself, which allows me to benchmark state-of-the-art proposals and to deduce some important design rules. Minimising the parasitic losses, for example, is of critical importance, as the desired scattering properties are directly proportional to these losses. To study the impact of parasitics, I quantify the useful absorption enhancement of two different light trapping mechanisms, i.e. diffractive vs plasmonic, based on a fair and simple experimental comparison. The experiment demonstrates that diffractive light-trapping is a better choice for photovoltaic applications, because plasmonic structures accumulate the parasitical losses by multiple interactions with the trapped light. The results of this thesis therefore highlight the importance of diffractive structures as an effective way of trapping more light in a thinner solar cell device, and will help to define guidelines for new designs that may overcome the 30% power conversion efficiency limit.

In recent years, the emergence of a new type of sub-wavelength dielectric grating with a high refractive index contrast, typically referred to as high contrast grating, and its simple fabrication procedure have given rise to numerous applications in integrated photonics and optoelectronics. The on-going demand for miniaturised planar devices in the near-infrared spectral range and the compatibility with standard processing techniques motivated us to explore the intriguing resonance effects in high contrast gratings for the application as broadband mirrors, reflective lenses, and as part of static and tuneable Fabry-Pérot cavities. We chose amorphous silicon as the high refractive index material. The characteristics of grating mirrors were analysed numerically and measured experimentally, confirming their suitability for use as focussing reflectors and cavity mirrors in a Fabry-Pérot interferometer. As a result, a new type of reflective diffractive lens incorporating multiple phase jumps was designed, fabricated, and experimentally tested, and showed improved performance in terms of numerical aperture compared to other devices in the literature. Furthermore, initial experiments by integrating arrays of such lenses with a microfluidic platform showed an application potential in the area of large-scale optical trapping. In an attempt to develop functional wavelength-selective filters utilising high contrast gratings and microelectromechanical systems, we investigated static cavities based on metals and gratings, which included simulation, development of the fabrication process, and optical characterisation. Following these proof-of-principle studies and the computational analysis of the mechanical behaviour of flexible membrane designs, the challenge in making tuneable devices was to fabricate a fully working tuneable filter and its characterisation. Overall, this thesis contributes to the field of integrated and planar silicon photonics by addressing the design and the performance aspects of high contrast grating lenses with high numerical aperture as well as addressing key challenges in the fabrication of tuneable filters.

Today's information and communication industry is confronted with a serious bottleneck due to the prohibitive energy consumption and limited transmission bandwidth of electrical interconnects. Silicon photonics offers an alternative by transferring data optically and thereby eliminating the restriction of electrical interconnects over distance and bandwidth. Due to the inherent advantage of using the same material as that used for the electronic circuitry, silicon photonics also promises high volume and low cost production plus the possibility of integration with electronics. In this thesis, I introduce an all-silicon optical interconnect architecture that promises very high integration density along with very low energy consumption. The basic building block of this architecture is a vertically coupled photonic crystal cavity-waveguide system. This vertically coupled system acts as a highly wavelength selective filter. By suitably designing the waveguide and the cavity, at resonance wavelength of the cavity, large drop in transmission can be achieved. By locally modulating the material index of the cavity electrically, the resonance wavelength of the cavity can be tuned to achieve modulation in the transmission of the waveguide. The detection scheme also utilizes the same vertically coupled system. By creating crystal defects in silicon in the cavity region, wavelength selective photodetection can be achieved. This unique vertical coupling scheme also allows us to cascade multiple modulators and detectors coupled to a single waveguide, thus offering huge channel scalability and design and fabrication simplicity. During this project, I have implemented this vertical coupling scheme to demonstrate modulation with extremely low operating energy (0.6 fJ/bit). Furthermore, I have demonstrated cascadeability and multichannel operation by using a comb laser as the source that simultaneously drives five channels. For photodetection, I have realized one of the smallest wavelength selective detector with responsivity of 0.108 A/W at 10 V reverse bias with a dark current of 9.4 nA. By cascading such detectors I have also demonstrated a two-channel demultiplexer.

Silicon Photonics has been a major success story in the last decade, with many photonic devices having been successfully demonstrated. The only missing component is the light source, however, as making an efficient light source in silicon is challenging due to the material's indirect bandgap. The development of a silicon light source would enable us to make an all-silicon chip, which would find many practical applications. The most notable among these applications are on-chip communications and sensing applications. In this PhD project, I have worked on enhancing silicon light emission by combining material processing and device engineering methods. Regarding materials processing, the emission level was increased by taking three routes. In all the three cases the emission was further enhanced by coupling it with a photonic crystal (PhC) cavity via Purcell effect. The three different approaches taken in this PhD project are listed below. 1. The first approach involves incorporation of optically active defects into the silicon lattice by hydrogen plasma treatment or ion implantation. This process results in broad luminescence bands centered at 1300 and 1500 nm. By coupling these emission bands with the photonic crystal cavity, I was able to demonstrate a narrowband silicon light emitting diode at room temperature. This silicon nano light emitting diode has a tunable emission line in the 1300-1600 nm range. 2. In the second approach, a narrow emission line at 1.28µm was created by carbon ion implantation, termed “G-line” emission. The possibility of enhancing the emission intensity of this line via the Purcell effect was investigated, but only with limited success. Different proposals for future work are presented in this regard. 3. The third approach is deposition of a thin film of an erbium disilicate on top of a PhC cavity. The erbium emission is enhanced by the PhC cavity. Using this method, an optically pumped light source emitting at 1.54 µm and operating at room temperature is demonstrated. A practical application of silicon light source developed in this project in gas sensing is also demonstrated. As a first step, I show refractive index sensing, which is a simple application for our source and demonstrates its capabilities, especially relating to the lack of fiber coupling schemes. I also discuss several proposals for extending applications into on-chip biological sensing.

Optical manipulation of nanoscale objects is studied with particular emphasis on the role of plasmon resonance for enhancement of optical forces. The thesis provides an introduction to plasmon resonance and its role in confinement of light to a sub-diffraction volume. The strong light confinement and related enhancement of optical forces is then theoretically studied for a special case of nanoantenna supporting plasmon resonances. The calculation of optical forces, based on the Maxwell stress tensor approach, reveals relatively weak optical forces for incident powers that are used in typical realisations of trapping with nanoantenna. The optical forces are so weak that other non-optical effects should be considered to explain the observed trapping. These effects include heating induced convection, thermoporesis and chemical binding. The thesis also studies the optical effects of plasmon resonances for a fundamentally different application - size-based optical sorting of gold nanoparticles. Here, the plasmon resonances are not utilised for sub-diffraction light confinement but rather for their ability to increase the apparent cross-section of the particles for their respective resonant sizes. Exploiting these resonances, we realise sorting in a system of two counter-propagating evanescent waves, each at different wavelength that selectively guide gold nanoparticles of different sizes in opposite directions. The method is experimentally demonstrated for bidirectional sorting of gold nanoparticles of either 150 or 130 nm in diameter from those of 100 nm in diameter within a mixture. We conclude the thesis with a numerical study of the optimal beam-shape for optical sorting applications. The developed theoretical framework, based on the force optical eigenmode method, is able to find an illumination of the back-focal plane of the objective such that the force difference between nanoparticles of various sizes in the sample plane is maximised.

The field of nanophotonics is a major research topic, as it offers potential solutions to important challenges, such as the creation of low power, high bandwidth interconnects or optical sensors. Within this field, resonant structures and slow light waveguides are used to improve device performance further. Photonic crystals are of particular interest, as they allow the fabrication of a wide variety of structures, including high Q-factor cavities and slow light waveguides. The high scattering loss of photonic crystal waveguides, caused by fabrication disorder, however, has so far proven to be the limiting factor for device applications. In this thesis, I present a detailed study of propagation loss in slow light photonic crystal waveguides. I examine the dependence of propagation loss on the group index, and on disorder, in more depth than previous work by other authors. I present a detailed study of the relative importance of different components of the propagation loss, as well as a calculation method for the average device properties. A new calculation method is introduced to study different device designs and to show that photonic crystal waveguide propagation loss can be reduced by device design alone. These “loss engineered” waveguides have been used to demonstrate the lowest loss photonic crystal based delay line (35 dB/ns) with further improvements being predicted (< 20 dB/ns). Novel fabrication techniques were investigated, with the aim of reducing fabrication disorder. Initial results showed the feasibility of a silicon anneal in a nitrogen atmosphere, however poor process control led to repeatability issues. The use of a slow-fast-slow light interface allowed for the fabrication of waveguides spanning multiple writefields of the electron-beam lithography tool, overcoming the problem of stitching errors. The slow-fast-slow light interfaces were combined with loss engineering waveguide designs, to achieve an order of magnitude reduction in the propagation loss compared to a W1 waveguide, with values as low as 130 dB/cm being achieved for a group index around 60.

In the past decade, great research effort was inspired by the need to realise active optical functionalities in silicon, in order to develop the full potential of silicon as a photonic platform. In this thesis we explore the possibility of achieving tunable delay and optical gain in silicon, taking advantage of the unique dispersion capabilities of photonic crystals. To achieve tunable optical delay, we adopt a wavelength conversion and group velocity dispersion approach in a miniaturised engineered slow light photonic crystal waveguide. Our scheme is equivalent to a two-step indirect photonic transition, involving an alteration of both the frequency and momentum of an optical pulse, where the former is modified by the adiabatic tuning possibilities enabled by slow light. We apply this concept in a demonstration of continuous tunability of the delay of pulses, and exploit the ultrafast nature of the tuning process to demonstrate manipulation of a single pulse in a train of two pulses. In order to address the propagation loss intrinsic to slow light structures, with a prospect for improving the performance of the tunable delay device, we also investigate the nonlinear effect of stimulated Raman scattering as a means of introducing optical gain in silicon. We study the influence of slowdown factors and pump-induced losses on the evolution of a signal beam along the waveguide, as well as the role of linear propagation loss and mode profile changes typical of realistic photonic crystal structures. We then describe the work conducted for the experimental demonstration of such effect and its enhancement due to slow light. Finally, as the Raman nonlinearity may become useful also in photonic crystal nanocavities, which confine light in very small volumes, we discuss the design and realisation of structures which satisfy the basic requirements on the resonant modes needed for improving Raman scattering.

“Lab on a Chip” technologies have revolutionized the field of bio-chemical analytics. The crucial role of optical techniques in this revolution resulted in the emergence of a field by itself, which is popularly termed as “optofluidics”. The miniaturization and integration of the optical parts in the majority of optofluidic devices however still remains a technical challenge. The works described in this thesis focuses on developing integration methods to combine various optical techniques with microfluidics in an alignment-free geometry, which could lead to the development of portable analytical devices, suitable for field applications. The integration approach was applied to implement an alignment-free optofluidic chip for optical chromatography; a passive optical fractionation technique fractionation for cells or colloids. This system was realized by embedding large mode area photonic crystal fiber into a microfluidic chip to achieve on-chip laser beam delivery. Another study on passive sorting envisages an optofluidic device for passive sorting of cells using an optical potential energy landscape, generated using an acousto-optic deflector based optical trapping system. On the analytical side, an optofluidic chip with fiber based microfluidic Raman spectroscopy was realized for bio-chemical analysis. A completely alignment-free optofluidic device was realized for rapid bio-chemical analysis in the first generation by embedding a novel split Raman probe into a microfluidic chip. The second generation development of this approach enabled further miniaturization into true microfluidic dimensions through a technique, termed Waveguide Confined Raman Spectroscopy (WCRS). The abilities of WCRS for online process monitoring in a microreactor and for probing microdroplets were explored. Further enhanced detection sensitivity of WCRS with the implementation of wavelength modulation based fluorescent suppression technique was demonstrated. WCRS based microfluidic devices can be an optofluidic analogue to fiber Raman probes when it comes to bio-chemical analysis. This allows faster chemical analysis with reduced required sample volume, without any special sample preparation stage which was demonstrated by analyzing and classifying various brands of Scotch whiskies using this device. The results from this study also show that, along with Raman spectroscopic information, WCRS picks up the fluorescence information as well, which might enhance the classification efficiency. A novel microfabrication method for fabricating polymer microlensed fibers is also discussed. The microlensed fiber, fabricated with this technique, was combined with a microfluidic gene delivery system to achieve an integrated system for optical transfection with localized gene delivery.

The growing speed and bandwidth requirements of telecommunication systems demand all-optical on-chip solutions. Microphotonic devices can deliver low power nonlinear signal processing solutions. This thesis looks at the slow light photonic crystals in chalcogenide glasses to enhance low power nonlinear operation. I demonstrate the development of new fabrication techniques for this delicate class of materials. Both, reactive ion etching and chemically assisted ion beam etching are investigated for high quality photonic crystal fabrication. A new resist-removal technique was developed for the chemical, mechanical and light sensitive thin films. I have developed a membraning method based on vapor phase etching in combination with the development of a save and economical etching tool that can be used for a variety of vapour phase processes. Dispersion engineered slow light photonic crystals in Ge₃₃As₁₂Se₅₅ are designed and fabricated. The demonstration of low losses down to 21±8dB/cm is a prerequisite for the successful demonstration of dispersion engineered slow light waveguides up to a group index of around n[subscript(g)] ≈ 40. The slow light waveguides are used to demonstrate highly efficient third harmonic generation and the first advantages of a pure chalcogenide system over the commonly used silicon. Ge₁₁.₅As₂₄24Se₆₄.₅ is used for the fabrication of photonic crystal cavities. Quality factors of up to 13000 are demonstrated. The low nonlinear losses have enabled the demonstration of second and third harmonic generation in those cavities with powers up to twice as high as possible in silicon. A computationally efficient model for designing coupled resonator bandpass filters is used to design bandpass filters. Single ring resonators are fabricated using a novel method to define the circular shape of the rings to improve the fabrication quality. The spectral responses of the ring resonators are used to determine the coupling coefficient needed for the design and fabrication of the bandpass filters. A flat top bandpass filter is fabricated and characterized as demonstration of this method. A passive all-optical regenerator is proposed, by integrating a slow-light photonic crystal waveguide with a band-pass filter based on coupled ring resonators. A route of designing the regenerator is proposed by first using the dispersion engineering results for nonlinear pulse propagation and then using the filter responses to calculate the nonlinear transfer function.

Nano-emitters are the new generation of laser devices. A photonic-crystal cavity, which highly confines light in small volumes, in combination with quantum-dots can enhance the efficiency and lower the threshold of this device. The practical realisation of a reliable, electrically pumped photonic-crystal laser at room-temperature is, however, challenging. In this project, a design for such a laser was established. Its properties are split up into electrical, optical and thermal tasks that are individually investigated via various device simulations. The resulting device performance showed that with our design the quantum-dots can be pumped in order to provide gain and to overcome the loss of the system. Threshold currents can be as low as 10’s of μA and Q-factors in the range of 1000’s. Gallium arsenide wafers were grown according to our specifications and their diode behaviour confirmed. Photonic-crystal cavities were fabricated through a newly developed process based on a TiOₓ hard-mask. Beside membraned cavities, also cavities on oxidised AlGaAs were fabricated with help to a unique hard-mask removal method. The cavities were measured with a self-made micro-photoluminescence setup with the highest Q-factor of 4000 for the membrane cavity and a remarkable 2200 for the oxide cavity. The fabrication steps, regarding the electrically pumped photonic-crystal laser, were developed and it was shown that this device can be fabricated. During this project, a novel type of gentle confinement cavity was developed, based on the adaption of the dispersion curve (DA cavity) of a photonic-crystal waveguide. Q-factors of as high as 600.000 were measured for these cavities made in Silicon.

Optical trapping of metal nanoparticles investigates phenomena at the interface of plasmonics and optical micromanipulation. This thesis combines ideas of optical properties of metals originating from solid state physics with force mechanism resulting from optical trapping. We explore the influence of the particle plasmon resonance of gold and silver nanospheres on their trapping properties. We aspire to predict the force mechanisms of resonant metal particles with sizes in the Mie regime, beyond the Rayleigh limit. Optical trapping of metal nanoparticles is still considered difficult, yet it provides an excellent tool to investigate their plasmonic properties away from any interface and offers opportunities to investigate interaction processes between light and nanoparticles. Due to their intrinsic plasmon resonance, metal nanoparticles show intriguing optical responses upon interaction with laser light. These differ greatly from the well-known bulk properties of the same material. A given metal nanoparticle may either be attracted or repelled by laser light, only depending on the wavelength of the latter. The optical forces acting on the particle depend directly on its polarisability and scattering cross section. These parameters vary drastically around the plasmon resonance and thus not only change the magnitude but also the direction and entire nature of the acting forces. We distinguish between red-detuned and blue-detuned trapping, that is using a trapping wavelength shorter or longer than the plasmon resonance of the particle. So far optical trapping of metal nanoparticles has focussed on a wavelength regime far from the particle’s resonance in the infrared. We experiment with laser wavelengths close to the plasmon resonance and expand the knowledge of metal nanoparticle trapping available to date. Existing theoretical models are put to the test when we compare these with our real experimental situations.

The aim of this project was to produce an integrate-able free space optical transceiver for Specks. Specks are tiny computing units that together can form a powerful network called a SpeckNet. The SpeckNet platform is developed by the SpeckNet consortium, which consists of five Scottish Universities and combines computer science, electrical engineering and digital signal processing groups. The principal goal of creating an optical transceiver was achieved by integrating in-house fabricated VCSELs (with lasing thresholds below 400 uA) and custom designed detectors on the SpeckNet platform. The transceiver has a very low power consumption (approximately 100 uW), which removes the need for synchronous communication through the SpeckNet thus making the network more efficient. I describe both static and dynamic beam control techniques. For static control, I used micro-lenses. I fabricated the lenses by greyscale electron beam lithography and integrated them directly on VCSEL arrays. I achieved a steering angle of 10 degrees with this design. I also looked at integrated gratings etched straight into a VCSEL and observed beam steering with an efficiency of 60% For dynamic control, I implemented a liquid crystal (LC) design. I built a LC cell with 30 individually controlled pixels, but I only achieved a steering angle of 1 degree. Furthermore, I investigated two different techniques for achieving beam steering by interference, using coupled VCSELs (a phased array approach). Firstly, using photonic crystals etched into the surface of the VCSEL, I built coupled laser cavities. Secondly, I designed and built bow-tie type VCSELs that were optically coupled but electrically isolated. These designs work by differential current injection causing an interference effect in the VCSELs far field. This technique is the first stepping stone towards realising a phased optical array. Finally, I considered signal detection. Using the same VCSEL material, I built a resonant-cavity detector. This detector had a better background rejection ratio than commercially available silicon devices.

The use of photonic crystals for light extraction from light-emitting diodes (LEDs) gives the possibility to shape the farfield emission pattern. This is of particular interest for étendue-limited LED applications that require a more directional farfield than state- of-the-art Lambertian emitters. However, the application of a photonic crystal in a LED results in directional emission only if the photonic crystal and the distribution of guided modes in the LED are tuned correctly. In this thesis, red- and blue-emitting thin-film PhC-LEDs in the AlGaInP and InGaN material systems were modelled, designed, fabricated and characterized. The first experimental results show that light extraction with photonic crystals from AlGaInP thin-film LEDs several microns thick is neither directional nor more efficient than state-of-the-art LEDs with a rough surface structure. Directional light extraction for AlGaInP PhC-LEDs is for the first time demonstrated in much thinner devices where the photonic crystal light extraction of guided modes is combined with the resonant-cavity effect. In an attempt to approach the ideal PhC-LED, strong photonic crystal farfield shaping is demonstrated in InGaN thin-film LEDs of sub-micron thickness. Analysis of their spectral farfields unexpectedly shows that high order diffraction contributes significantly to the light extraction efficiency if the mode absorption is sufficiently low. It is also demonstrated that directional photonic crystal light extraction is possible in InGaN thin-film LEDs several microns thick. The directionality stems from the modulation of the spontaneous emission caused by the proximity of the active region to the bottom mirror. Two new concepts for enhanced light extraction and high directionality are presented: Photonic crystals with two dominating lattice constants are found to outperform conventional photonic crystal LEDs. An alternative approach is the dielectric PhC-LED - FDTD simulations show that the high extraction efficiency of LEDs with surface roughness is combined with the higher directionality of photonic crystal light extraction.

The research described in this thesis relates to the design, fabrication and testing of novel semiconductor-based light sources that have been designed for the generation of infra-red light. The thesis is formatted to account for two distinct components of my work, where the first part concerns sources producing coherent light by direct laser emission, notably, ultrashort-pulse quantum-dot lasers. These types of lasers continue to show considerable promise as efficient, compact sources of ultrashort pulses with durations of hundreds of femtoseconds, while giving rise to unique and interesting electronic properties such as low lasing thresholds through the quantum nature of their density of states. At the outset a study of the most relevant aspects of the lasing dynamics of an optically pumped quantum-dot laser is outlined. Pumping of the device with intense discrete optical pulses leads to output from multiple electronic states, each having a characteristic wavelength and temporal properties. I show that pulses produced by excited-state emission have shorter durations (24 ps) and arrive earlier in time than those due to transitions from the ground state, which themselves have durations of around 180 ps. Investigations are then made on two different mode-locked quantum-dot laser systems. One is an all-quantum-dot external-cavity laser that is mode locked using a quantum-dot SESAM device at a repetition frequency of 860 MHz with output power approaching 20 mW. This is followed by a study of a monolithic two-section quantum-dot laser that is mode locked stably in a wide temperature range of 20°C to 70°C. The excellent performance characteristics presented serve to demonstrate both the versatility of quantum-dot material as components in mode-locked laser systems and the temperature stability of such laser devices. The second part of the thesis relates to structures that are designed to take advantage of nonlinear frequency conversion in GaAs-based semiconductors. This material system possesses a nonlinear coefficient of ~170 pm/V and is transparent from around 0.9 μm through to 17 μm, making it attractive for the realisation of a new class of efficient, integrable, quasi-phase-matched, optical parametric oscillator devices. Initially, ion implantation is utilised as a vector to create a periodically-switched nonlinear ridge waveguided device. The observation is made that in the course of implantation the transmissive properties of the device are severely degraded. Unfortunately, the high losses incurred, which reached 250 dB/cm, could not be removed without also destroying the modulation in nonlinearity. During the course of this investigation, significant technological advances were made in the production of orientation-patterned GaAs structures. By recognising the elegance and potential of this new orientation-patterned (OP) methodology, a study of its implications and applicability in the context of my project is initiated.

In recent years, interest has grown in the research fields of semiconductor lasers and photonic crystals. This thesis looks at integrating photonic crystals into existing semiconductor laser technology to act as functional laser mirrors. The majority of the research is conducted on a quantum-dot material system. The surface recombination velocity of a GaAs based quantum-dot material is shown to be a similar value to InP material. This allows the creation of fine photonic crystal structures in the laser design without high threshold current penalties. The spectral reflection properties of a one dimensional photonic crystal is studied and found to be an unsuitable candidate for a stand-alone laser mirror, due to its low reflectivity. A two-dimensional photonic crystal W3 defect waveguide is successfully integrated as a quantum-dot laser mirror. Single fundamental mode output is achieved with a typically multi-mode 20 μm wide laser mesa, highlighting the mode selective property of the mirror. A similar two-dimensional mirror is studied for its potential as a dispersion compensating mirror for mode-locked lasers. Initial theoretical analysis shows pulse compression for a suitably designed mirror. Experimental continuous- wave results for the same mirror structure demonstrate the tuning of mirror reflectivity with photonic crystal hole radius. A hybrid silicon-organic photonic crystal laser is demonstrated with output in the visible spectrum. This design is a new type of silicon emitter.

Three projects are described in this thesis that combine microfabrication techniques with optical micromanipulation. The aim of these projects is to use expertise in microlithography and optical tweezing to create new tools for Lab-on-Chip devices. The first project looks at the creation of microgears that can be moved using an optical force. The microgears include one dimensional photonic crystal that creates birefringence. This allows the transfer of angular momentum from a circularly polarised light beam to the microgear, making them spin. The microgears are simulated, fabricated and tested. Possible biological applications are suggested. The second project looks at creating microchannels to perform micromanipulation experiments in. Different methods of fabricating the microfluidic channels are compared, and the resulting chambers are used to find the maximum flow rate an optical sorting experiment can be performed at. The third project involves using a thin photoconductive layer to allow the optical control of an electrical force called dielectrophoresis. This light induced dielectrophoresis (LIDEP) allows similar control to optical tweezing but requires less irradiance than optical tweezing, allowing control over a larger area with the same input optical power. A LIDEP device is created and experiments to measure the electrical trap size that is created with a given optical spot size are performed. These three projects show different microfabrication techniques, and highlight how well suited they are for use in optical manipulation and microfluidic experiments. As the size of objects that can be optically manipulated matches well with the size of objects that can be created with microfabrication, it seems likely that many more interesting applications will develop.

Photonic Crystal structures have been heralded as a disruptive technology for the miniaturization of opto-electronic devices, offering as they do the possibility of guiding and manipulating light in sub-micron scale waveguides. Applications of photonic crystal guiding - the ability to send light around sharp bends or compactly split signals into two or more channels have attracted a great deal of attention. Other effects of this waveguiding mechanism have become apparent, and attracted much interest - the novel dispersion surfaces of photonic crystal structures allow the possibility of “slow light” in a dielectric medium, which as well as the possibility of compact optical delay lines may allow enhanced light-matter interaction, and hence miniaturisation of active optical devices. I also consider a third, more traditional type of photonic crystal, in the form of a grating for surface coupling. In this thesis, I address many of the aspects of passive photonic crystals, from the underlying theory through applied device modelling, fabrication concerns and experimental results and analysis. Further, for the devices studied, I consider both the relative merits of the photonic crystal approach and of my work compared to that of others in the field. Thus, the complete spectrum of photonic crystal devices is covered. With regard to specific results, the highlights of the work contained in this thesis are as follows: Realisation of surface grating couplers in a novel material system demonstrating some of the highest reported fibre coupling efficiencies. Development of a short “injecting” taper for coupling into photonic crystal devices. Optimisation and experimental validation of photonic crystal routing elements (Y-splitter and bend). Exploration of interfaces and coupling for “slow light” photonic crystals.